Research on the Application of the Transparency Principle in the EU Artificial Intelligence Act

Abstract

The EU Artificial Intelligence Act establishes the principle of transparency to ensure the traceability, interpretability, and communicability of artificial intelligence (AI) systems. This principle is reflected to varying degrees in AI systems of different risk levels, with particularly stringent requirements for high-risk AI systems. However, in practical application, the transparency principle faces multiple challenges, such as conflicts with trade secret protection, technical complexity, and ambiguous liability definition. Although the EU has adopted a series of specific provisions to ensure the implementation of the transparency principle, including requirements for information disclosure and documentation, these measures still need to further balance the interests of all parties during implementation to achieve the goal of promoting the healthy development of AI technology while protecting public rights and interests, social fairness, and justice.